AD Biomarkers Tied to Worse Cognition in Midlife Adults

Evidence of amyloid and tau pathology was associated with worse cognitive performance and accelerated cognitive decline in midlife adults without dementia, a new cohort study showed. Positivity for the Alzheimer’s disease (AD) blood biomarkers was rare, reported in only 6%-15% of participants. However, the cognitive impacts were significant, with lower performance in processing speed and executive function and a faster decline in verbal memory and processing speed over 5 years than those without biomarker-defined AD neuropathology. Researchers said the study, published recently in The Lancet , is among the first to examine early AD neuropathology biomarkers and their associations with cognition in midlife. Its findings suggest that biological changes associated with AD may be tied to measurable differences in cognition decades before dementia onset, said senior investigator Kristine Yaffe, MD, director of the Center for Population Brain Health at the University of California, San Francisco. “Alzheimer’s disease pathology begins years before symptoms emerge,” Yaffe said in a news release . “Detecting the disease early means patients can target modifiable risk factors and maybe seek other care.”

Detecting Preclinical AD Changes

Amyloid-beta accumulation and tau pathology begin well before clinical symptoms and can be detected with cerebrospinal fluid or PET imaging, although these methods are invasive, costly, or not feasible for use in large-scale population screening. Blood biomarkers, particularly phosphorylated tau 217 (p-tau217) and amyloid-beta ratios, are a promising minimally invasive alternative for detecting AD neuropathology. Regulatory interest in these assays has also grown, including FDA clearance of a plasma p-tau217/amyloid-beta-42 assay for use in symptomatic individuals. However, prior studies have focused on older adults or memory clinic populations, leaving it unclear how these biomarkers perform in cognitively unimpaired adults in midlife and in more diverse community-based cohorts. To evaluate these biomarkers in midlife, investigators included 1350 participants (median age, 61 years; 58% women; 45% Black; 55% White) from the CARDIA study, a long-running community cohort of Black and White adults from four US cities. Participants were enrolled between the ages of 18 and 30 years and completed follow-up examinations every 2-5 years over a 35-year period. Vascular risk factors and comorbidities were common, including hypertension in about 60% of participants. At year 30 of the study, participants underwent cognitive testing to assess global cognition, processing speed, executive function, verbal memory, and verbal fluency. Five years later, when participants were aged 53-69 years, they underwent cognitive testing again and provided blood samples for biomarker measurements of p-tau217, amyloid-beta-42, and amyloid-beta-40. Accelerated decline was defined as a 5-year decline of ≥ 1.5 SDs greater than the cohort mean change in each cognitive domain. Analyses were adjusted for demographic factors, education, BMI, renal function, and APOE epsilon 4 status.

Biomarkers Signal Cognitive Decline

Among all participants, 6% were positive for AD neuropathology, 86% were negative, and 8% were intermediate, meaning their biomarker levels were not high enough to classify as AD neuropathology, but not low enough to rule it out. Positivity rates varied depending on biomarker, reaching 15% for amyloid-beta-42/40 and 4% for p-tau217 alone. Participants with biomarker-defined AD neuropathology were more likely to carry the APOE epsilon 4 allele ( P < .0001). Baseline cognitive performance was significantly worse in individuals with elevated levels of AD biomarkers compared to those without biomarker-defined AD neuropathology. For processing speed, adjusted standardized differences in cognitive scores ranged from -0.54 in p-tau217-positive individuals to -0.25 in those with amyloid-beta-42/40, with P values ranging from.0001 to.0048. For those with both p-tau217 and amyloid-beta-42, the difference was -0.34. Executive function showed a similar pattern, with adjusted standardized differences ranging from -0.42 in those with p-tau217 to -0.19 in those with amyloid-beta-42/40, with a difference of -0.27 for individuals with p-tau217 and amyloid-beta-42 ( P ≤ .049 for all). No significant associations were observed for global cognition or verbal fluency. When tested 5 years later, those with AD pathology had significantly greater odds of accelerated decline over 5 years on verbal memory (amyloid-beta-42/40 odds ratio [OR], 4.31; p-tau217/amyloid-beta-42 OR, 2.44) and processing speed (p-tau217 OR, 3.98; p-tau217/amyloid-beta-42 OR, 3.35) than those with no amyloid or tau. Some effect modification was observed, with stronger associations between biomarkers and cognition among women, Black participants, and APOE epsilon 4 carriers. The association between p-tau217 and processing speed was stronger in Black participants than in White participants ( P for interaction = .0396), while associations involving executive function differed by sex and APOE status in some models.

Current Clinical Use Limited

Although the findings suggest a biological signal of blood biomarkers in midlife, Yaffe emphasized that clinical use remains limited in asymptomatic adults. “With a low prevalence, one has to be careful for false positives,” she told Medscape Medical News . “Today, clinicians should restrict the use of blood AD biomarkers to those they are concerned have clinical cognitive issues to help decide if it’s AD or another process.” Still, she suggested the assays may soon become part of routine frontline evaluation in symptomatic patients. “I think we are very close to using blood biomarkers as first-line in diagnosis, but mostly for those with symptoms,” Yaffe said. Negative results could help rule out AD, while positive findings may guide further confirmatory testing and treatment decisions, she added.

More Biomarker Research Needed

The study adds much-needed data on cognitively unimpaired, community-dwelling midlife populations, an area where evidence has been limited, Anna Rosenberg, PhD, and Tiia Ngandu, MD, PhD, wrote in an accompanying editorial . The racially diverse cohort, which included 45% Black participants, is a key strength given the ongoing underrepresentation in AD research, noted the authors, both of the Finnish Institute for Health and Welfare in Helsinki, Finland. However, they cautioned that in low pretest probability settings, such as asymptomatic adults, blood biomarkers could increase the number of false-positive results and “should not be used as a stand-alone indicator without other biological or clinical information.” The lack of long-term cognitive data and single-timepoint biomarker analyses in the study are key limitations to consider, they added, noting that the “performance of these biomarkers in cognitively unimpaired, middle-aged populations remains unclear.” Looking ahead, future studies should track biomarker trajectories over time and tie them to clinical outcomes, they wrote. “We still need a clearer understanding of the biological processes that each blood biomarker reflects,” suggesting broader markers capturing inflammation, neurodegeneration, and synaptic dysregulation should be included. Disclosure information for study authors is available in the original study publication. Rosenberg and Ngandu reported having no relevant financial disclosures.